BIOMAG 2000 CD-ROM

Current sources of initial QRS forces in left ventricular hypertrophy, WPW
syndrome, and left bundle branch block - origin of septal vector

Y. Nakaya/’^ M. Nomura,^ and H. Miyajima^

^Department of Nutrition and Metabolism, and ^Second Department of Medicine,
the University of Tokushima, School of Medicine, Tokushima, Japan 770-8503

1 Introduction

Since the original investigations of Lewis and
Rothsenchild (1915) [1], septal activation has been
thought to occur first on the left septal surface and
thereafter to spread from the left side to the right
side of the intraventricular septum. Thus, the initial
QRS vector in electrocardiogram is considered to be
due to activation of intraventricular septum and
called as septal vector. At this time (within 20 msec
from the onset of the QRS wave), posterobasal area
is not activated. However, normal septal vector
disappears in cases of inferior infarction as well as
anteroseptal infarction. Then arises the question
whether the initial QRS vector is really composed of
intraventricular septum alone. In left ventricular
hypertrophy, the initial QRS vector becomes smaller
or even absent. Although the mechanism of
disappearance of septal vector in left ventricular
hypertrophy is now considered to be due to septal
fibrosis, the precise mechanism remains unclear [2].
Therefore, we studied the origin of initial QRS force
by magnetocardiogarrm (MCG). The MCG three-
dimensional (3-D) source localization using
superconducting quantum interference device
(SQUID) gradiometer has an excellent spatial and
temporal resolution. To clarify the mechanism of
disappearance of septal vector (QS pattern in lead
Vi) in LVH, the MCG of normal subjects and the
patients with left ventricular hypertrophy were
recorded in an rf-shielded room using biomagnetic
measuring system. We also studied the initial QRS
force in patients with left bundle branch block
(LBBB) and type B WPW syndrome, in which the
septal vector is absent. The results of the study
provide a new interpretation of ECG wave forms.

2 Methods
2.1 Subjects

We studied 20 normal subjects (N group), 33
patients with left ventricular overload (LVH group),
10 patients with complete LBBB (LBBB group) and
12 patients with type B WPW syndrome (WPW
group), in which initial QRS vector is also
frequently deviated. The LVH group was subdivided

into two subgroups according to the presence or
absence of an initial r waves in lead Vi, so-called
septal vector, i.e. those with rS pattern (LVH-rS
group) and those without r wave (LVH-QS group).
In the LVH-QS group, the r wave in lead Vi and/or
q wave in lead V6 was absent. The LVH-rS group
shows the r wave in lead Vi and q wave in lead V6.
The echocardiogram and magnetic resonace imaging
(MRI) were recorded in all subjects to localize
anatomically current sources.

2.2 Construction of isofield map and source
localization

The MCG was recorded over the anterior chest wall
with a second-derivative superconducting quantum
interference device (SQUID) gradiometer (single
channel rf SQUID; Model BMP, BTI, San Diego, or
7-ch dc-SQUID, Liquid Gas Corp. Osaka) in an rf-
shielded room at Tokushima University Hospital.
The patients were placed in the supine position on a
nonmagntic bet. All metal objects, including
watches, metallic contents of trousers, such as keys
and hairpins, were removed. The magnetic field
perpendicular to the anterior chest wall was
measured. The MCG recoding was taken by
scanning each recording point (25 points or 63
points) with a bandpass filter of 0.02 - 300Hz. The
position of detector was confirmed by laser point
indicator. The reference points used were the
xyphoid process and the top of the sternum. The
MCG measurements were based on a three-
dimensional (3-D) coordinate system consisting of a
median line between the xyphoid process and the
top of the sternum. The origin was defined as the
xyphoid process. The Y axis was defined as the line
passing up from the origin and parallel to the sagital
plane (inferior to superior); the X as the line passing
out of the left side from the origin (right to left), and
the Z axis as the line passing from the origin and
perpendicular from the plane of the detection coil
(anterior to posterior).

Isofield map at every 2 msec was constructed by
using the same methods as construction of
isopotential map [3]. In order to localize the current

3 Results

Figure 1: Isofield map and vector arrow map at
20msec in a normal subject. Initial QRS vector
directed to the right and inferiorly.

dipole, a single dipole model [4,5] was used and the
3-D loeation of the equivalent eurrent dipole was
eomputed every 2ms from the onset of the QRS
eomplex by the least square method. The 3-
dimentional dipole loeations were superimposed
onto the MRI of the individual subjeets providing
anatomieal loealization in the ventriele.

2.3 Vector arrow map

We also eonstrueted veetor arrow maps to study the
multiple instantaneous veetor aeeording to the
methods by Hosaka and Cohen [6]. A veetor arrow
(A) was defined as follows:

A = (3Bn/9y)x - (3Bn/9x)y

where x and y are axes of the body, x in the
direetion from right to left, and y from head to foot,
and X and y are veetors of unit length along the x
and y axes. Veetor arrow was obtained at eaeh point.
Veetor arrows on the map are supposed to indieate
the underlying eurrent pattern parallel to the frontal
plane.

3.1 Initial QRS wave in normal subjects

Fig. 1 shows isofield map and veetor arrow map in a
normal subjeet. The dedueed dipole of normal
subjeets was loeated on the ventrieular septum in 9
of 10 subjeets who reeorded MRI. Mean values of
dedueed dipole of the initial 20 msee veetor were
48+18 mm left ward, 58+24 mm superior to the
xyphoid proeess and 41+13 mm deep from the
surfaee of the gradiometer. The dedueed dipole at
early QRS moved from mid-septum to the apieal
septum in most of the eases.

(A) (B)

Figure 2; Single moving dipole in a normal subject
(A), a patient with essential hypertension (EH) and a
patient with aortic valve disease (ASR). Note that
20 msec dipole (2) directed to the left and displaced
to the left in ASR.

Table 1: Location of the single dipole at 20 msec in
normal subjects and patients with left ventricular
hypertrophy.

Group

n

Distanee from eoil

Normal

32

56+2 mm

LVH-rS

25

62+4 mm **

LVH-QS

13

70+5 mm**

WPW

12

54+9 mm

LBBB

10

53+5 mm

* p<0.05, p<0.01 vs Normal

WPW-B lBBB Normal

LVH(rS)

LVH(QS)

Figure 3: Location of deduced dipole in each group
on MRI.

3.2 Initial QRS wave in patients

Fig. 2 shows the movement of single dipoles during
ventrieular depolarization in normal subjeet and
patients with LVH. In patients with LVH (ARS)
initial QRS dipole was displaeed to the left and its
direetion was also displaeed to the left. At SOmsee
further inerease in veetors direeting leftward was
observed. Table 1 shows that the dedueed eurrent
souree in LVH group was displaeed posteriorly and
to the left. The deviation was signifieantly greater in
the LVH-QS group than that of LVH-rS group.
However, those of LBBB and WPW groups were
not displaeed and loeated similarly with that of
normal subjeets. Fig. 3 shows the MRI in whieh the
dipole dedueed by isofield mapping in eaeh group
was superimposed.

3.3 Analysis by vector arrow map

Fig. 4 shows the ECG, isofield map and veetor
arrow map of a patient with left ventrieular overload
with small r wave in lead Vi but not q waves in left
preeordial leads (V 5 and ¥ 5 ). At 20 msee Veetor
arrow map shows that veetor arrows loeating in the
right shows normal veetor direeting rightward but
those loeating in the left direeting to the left with
inereased amplitude. The dipole dedueed by isofield
map was direeted to the left probably as a result of
the summation of veetor arrows of both direetions.
Compared to this patient the veetor arrow map in a
patient with LBBB (Fig.5) showed dipoles direeting
to the left was not inereased in amplitude at the left
part. In late phase of depolarization of LBBB, the
inerease of veetor arrows in amplitude were not
observed.

3 Discussion

In the present study, we determined the loeation of
the initial QRS foree in various pathologieal
eonditions in whieh the septal veetor is absent. The
present study shows that the initial QRS foree
dedueed by the MCG was displaeed posteriorly in
left ventrieular hypertrophy and was inereased in
magnitude, whieh were different from that of LBBB
or type B WPW sundrome. These results suggest
that absenee of septal veetor might be due to the
inereased eleetromotive forees of the hypertrophied
left ventriele rather than eonduetion disturbanee.
The results of the present studies explain the
meehanism of abnormal initial QRS forees in ECG.

The initial QRS foree has been eonsidered to
originate from the aetivation of the left-sided septum
whieh moves to the rightward. It is eonsidered that
posterobasal area of the left ventriele is aetivated
later than septum and that this part does not
eontribute to generation of normal septal veetor.
However, we have reported that in inferior
myoeardial infaretion initial QRS veetor was

ECG

I U lU

J J

Firure 4: Isofield map and vector arrow map of a
patient with left ventricular overload with small r
wave in lead In the vector arrow map at 20 msec,
dipoles located on the left are increased in
amplitude and directed to the left.

■ • ■ ^ ^

• AT ^

■ / / ^ -

^ ^

* \

Normal LBBB

Figure 5: Vector arrow map in a normal and a
patient with LBBB. Note that there is no increase in
vector arrow in the left part of the map in
comparison with the map in figure 4.

displaced, suggesting that posterobasal area
(infarct area) also contributes initial QRS force.
These findings strongly support the idea that the
initial QRS vector is not originated from ventricular
septum alone, and that many areas of ventricle
contribute the formation of initial QRS vector.

In left ventricular hypertrophy, the initial QRS
vector is frequently deviated, which was explained
by septal fibrosis due to hypertrophy. The vector
arrow map studies also suggested that the rightward
vector directed to the left suggesting normal
rightward electromotive force, although decreased in
amplitude. In addition to the normal septal vector,
there were increased leftward directing
electromotive force in the left ventricle. Thus, the
resultant QRS force as a summation of many
portions of the both ventricles is located leftward
and directed to the left, because the balance of the
electromotive force between the left and right
ventricles was altered. (Fig. 6)

Figure 6: The mechanism of QS pattern in lead Vi in
LVH (hypothesis from this study). The initial QRS
force is a summation vector of various portion of the
ventricles. In LVH, hypertrophied left ventricle
produces larger electromotive force directed to the
left and posteriorly, resulting in displacement of the
initial vector to the left and posteriorly.

The initial QRS vector has been considered to be
due to the intraventricular activation. However, the
present study shows that other parts of the ventricle
also contribute to generate initial QRS vectors. Thus

the initial QRS vector is a hypothetical resultant or
mean vector representing the average direction and
magnitude of all electric forces produced at this
time. In agreement with our studies, three
endocardial areas are synchronously depolarized
from 0-5 msec after the onset of ventricular
activation in the studies of human heart [7]. These
studies also support the results of the present studies
that the initial QRS vector is summation of
electromotive forces of many areas of the both
ventricles.

References

[1] T. Lewis, and M.A. Rothschild, ‘The excitatory
process in the dog’s heart”, Philos. Trans.
Royal Society of London, 206: 181-189, 1915.

[2] W.E. Gaum, T.C. Chou, and S. Kaplan, “The
vectorcardiogram and elelctrocardigram in
supravalvular aortic stenosis and coarctation of
the aorta” Heart J., 84: 620-628, 1972.

[3] Y. Nakaya, M. Sumi, K. Saito, K. Fujino, M.
Murakami, and H. Mori, “Analysis of current
sources of the heart using isomagnetic and
vector arrow maps”, Jpn. Heart J. 25: 701-711,
1984.

[4] G.L. Romani, S.J. Williamson, L. Kaufman,
“Biomagnetic instrumentation”. Rev. Sci.
Instrum. 53: 1815-1845,1987

[5] S.J. Williamson, and L. Kaufman, “Magnetic
fields of the cerebral cortex”. In Biomagnetism
Erne S, Hahlbohm HD, Lubbig H, Walter de
gruyter, Eds. Berlin, New York, pp.354-402,
1981

[6] H. Hosaka, and D. Cohen, “Part IV: visual
determination of generators of the MCG”, J.
Electrocardiol. 9:426-432, 1976.

[7] D. Durrer, R.T. van Dam, G.E. Freud, M.J.
Janse, F.L. Meijler, and R.C. Arzbaecher,
“Total excitation of the human heart”.
Circulation, 41: 899-912, 1970.